Lecture: Knowledge II: What Knowledge Is
Lecture script · about 15–20 minutes spoken · Week 5
Opening
Picture a fork in a hiking trail. A wooden sign says "Lake" and points left, so you go left, and there's the lake. Here's the catch. A storm had knocked the sign loose, and it spun on its post and stopped, by chance, pointing the right way. You believed it, you had a good reason, and you were right. Did you know?
That question isn't only for hikers. A nursing student deciding when a chart reading counts as knowing a person's condition is asking it. So is a computer science major trusting a status light on a server, or a juror asked whether a witness really knew what they saw. In each case, being right by luck doesn't feel good enough.
In Week 1 you learned to test a definition with a counterexample. In Week 2 you learned standard form and validity. In Week 3, Plato split appearance from reality, and in Week 4, Descartes searched for certainty while Locke and Hume rested knowledge on experience. In this lecture I'll give you the most famous definition of knowledge, and then the cases built to break it. After that we'll ask where knowledge comes from, and what good reason even is. Next week, in Week 6, the course turns to the mind.
1. Justified true belief
Let's start with what knowing seems to take. Think of a nurse who says, "I know the person in bed 4 is allergic to penicillin." For that to be knowing, three things seem to matter. The person really has to be allergic. The nurse has to believe it. And the nurse needs a good reason, like a confirmed chart entry, rather than a hunch.
That gives us the classic definition. Justified true belief is the traditional analysis of knowledge: a person knows a claim when the claim is true, the person believes it, and the person is justified in believing it. A belief is a person's taking a claim to be true. Justification is good reason or support for holding a belief, such as evidence, a reliable source, or sound reasoning.
Each condition has a job. You can't know something false, however sure you feel. You can't know something you don't believe. And a lucky guess isn't knowing.
Pause and try this. A fan is certain their team will win, and the team wins. Which condition of justified true belief is missing?
Run through the three conditions one at a time.
Here's the answer. Justification is missing. The belief was held and turned out true, but feeling certain about a game isn't a good reason, so the fan guessed right rather than knew.
2. Good reason, not certainty
Last week left a worry hanging. Descartes wanted a belief no doubt could touch. If knowing needs that, you know very little.
Dataset B, the course's argument set, has an argument for exactly that. It's illustrative, written for teaching. If knowledge requires certainty, then I know very little. Knowledge requires certainty. So I know very little. That's modus ponens, so it's valid.
Now look back at the definition. It asks for justification, and justification can fall short of certainty. The nurse's chart could, in some odd case, be wrong, and on this definition the nurse can still know. So a defender of justified true belief rejects the second premise. Whether knowing needs certainty is still an open question, so the course keys this argument as valid and never as sound.
Pause and try this. Does the justified true belief definition require you to be certain?
Read the third condition word for word.
Here's the answer. No. It requires that you be justified, which means having good reason, and good reason can fall short of certainty.
3. One counterexample is enough
A definition is a bold thing. It doesn't say "usually." It says "every." Two terms help here. A necessary condition is a condition that must hold for something to be the case. A sufficient condition is one whose holding is enough, on its own, for something to be the case.
The classic definition treats truth, belief, and justification as each necessary, and all three together as sufficient. So it claims that every justified true belief is knowledge.
Here's a small story. A coach has a rule: a player is ready for the match if and only if they finished every practice. Then a player finishes every practice and sprains an ankle the night before. That one player breaks the rule. You don't need a pile of cases. A claim about every case falls the moment one case fails.
Pause and try this. How many cases does it take to defeat the claim that every justified true belief is knowledge?
Think about what "every" commits you to.
Here's the answer. It takes 1. A single justified true belief that isn't knowledge shows the definition, as stated, is false.
4. The Gettier cases
Here's the case at the center of this week. In 1963, Edmund Gettier published a paper called "Is Justified True Belief Knowledge?" in the journal Analysis. It runs three pages. In it he described cases where someone's belief is justified and true, but true only by luck, so the person seems not to know.
We call these Gettier cases. A Gettier case is a case in which a person has a belief that is justified and true, but true only by luck, so that the person seems not to know. The case's question is: is justified true belief enough for knowledge? It tests the analysis of knowledge.
Put a date beside it. Descartes' Meditations came out in 1641, and Gettier's paper came 322 years later. Descartes asked whether anything can be known for sure. Gettier asked whether the standard definition of knowing is even right.
One note on sources. I won't retell Gettier's own cases in detail. The cases you'll hear in this lecture were written for this course on the same shape.
Pause and try this. True or false: justified true belief is the settled definition of knowledge that philosophers agree on.
Think about why Gettier's paper is still discussed.
Here's the answer. False. It's a famous definition, not a settled one, and the Gettier cases are the reason philosophers still debate what knowledge is.
5. The trail sign, step by step
Back to the trail, which is a variation written for this course. Check the three conditions. Is it true that the left trail leads to the lake? Yes. Does the hiker believe it? Yes. Is the hiker justified? Yes, because a posted sign is good evidence. That's three out of three.
Yet the belief is right by luck. The storm-spun sign would have pointed left whichever way the lake was. What made the belief true, the land itself, had nothing to do with what justified it, the sign.
Now the argument. If every justified true belief is knowledge, then the hiker knows the left trail leads to the lake. The hiker doesn't know that. So not every justified true belief is knowledge. If this, then that; not that; so not this. That's denying the consequent, the same form as argument 2 in Dataset B, and it's valid.
The second premise is an intuition about a thought experiment, and some philosophers resist it. So the course keys the argument's validity, never the answer to the case. If you feel the hiker did know, say so and defend it.
Pause and try this. Name the form of the hiker argument, and say whether it's valid.
Look at what the second premise denies.
Here's the answer. It's denying the consequent, and it's valid, because the second premise denies the "then" part of the first.
6. The shared shape
Once you see the shape, it shows up all over. A student checks homework against a careful study partner's. The partner made two slips that cancel and still landed on 200, the right total. A help-desk worker tells a caller the server is running because a status light is green, but the light has been stuck for a week. The server happens to be running.
Each belief is true, believed, and justified. And each one is right by luck. Here's the thread through all of them: what justifies the belief isn't what makes it true. The partner's track record justified the student, but correct arithmetic, which the partner didn't do, made the total right. The light justified the worker, but the server's real state made the belief true.
Pause and try this. If the server had been down and the worker still said it was running, would that be a Gettier case?
Check the first condition.
Here's the answer. No. The belief would be false, and the definition already says a false belief isn't knowledge, so a Gettier case needs the belief to come out true.
7. A priori and a posteriori
Here's a different question about knowledge: where does it come from? Take "every triangle has 3 sides." You could check it in a dark room, with no map and no window. Now take "the left trail leads to the lake." Only experience can tell you that.
A priori means knowable independently of experience, as in mathematics; it describes how a claim is known, not when a person first learned it. A posteriori means knowable only through experience, such as what the senses report about the world. Last week's rationalists and empiricists disagree about how much of our knowledge of the world is a priori.
The word "prior" fools people. It sounds like "learned first." But someone who first meets a math fact late in life still knows it a priori, because the knowing doesn't lean on experience of the world.
Pause and try this. Is "the left trail leads to the lake" a priori or a posteriori?
Ask how anyone could find it out.
Here's the answer. It's a posteriori. Nothing in the words tells you where the trail goes, so only experience can.
8. The regress problem
A child asks a parent, "Why do you think it'll rain?" "The forecast says so." "Why trust the forecast?" "It's usually right." "How do you know that?" Every reply is another belief, and the child can always ask again.
That game points at a real problem. The regress problem is the problem that if every justified belief needs another belief to justify it, the chain of reasons seems to go on forever, loop back on itself, or stop at something unsupported.
A skeptic is a person who argues that we know far less than we think, or that our beliefs lack the justification that knowledge needs. The skeptic presses hard here. An endless chain never finishes. A loop props a belief up on itself. And a chain that stops seems to rest on nothing. So, says the skeptic, nothing is justified.
Pause and try this. Name the three ways a chain of reasons could go.
Think about the child's game of "why."
Here's the answer. It could go on forever, loop back on itself, or stop at something with no further reason behind it.
9. Three answers to the regress
This week compares three answers, each stated the way its defenders would.
Foundationalism is the theory of justification that holds that some beliefs are basic and support the rest, so the chain of reasons comes to an end. Think of a building on a foundation.
Coherentism is the theory of justification that holds that beliefs are justified by fitting together in a system of beliefs that support one another, so no belief has to come first. Think of a jigsaw puzzle, where each piece is held by how it fits.
Reliabilism is the theory of justification that holds that a belief is justified when it is produced by a reliable process, one that tends to produce true beliefs. Think of a thermometer you trust because it tends to read right.
Picture a lifeguard who believes a swimmer is waving for help. A foundationalist says it rests on a basic belief about what the lifeguard sees right now. A coherentist says it fits everything else the lifeguard believes about the water. A reliabilist says it came from trained eyesight in daylight.
Pause and try this. Which theory says a belief can be justified with no chain of reasons the person could recite?
Think about the thermometer.
Here's the answer. Reliabilism. It says justification depends on how the belief was formed, by a reliable process, not on reasons you can list.
10. Testing each answer
Each answer meets a strong objection, and each has a reply. Foundationalism first. If nothing supports a basic belief, stopping there looks arbitrary. The reply: a basic belief is supported by experience or by being self-evident, not by another belief, and support from experience is still support, so the stopping point isn't arbitrary.
Coherentism next. A made-up story can fit together perfectly and still be false. The reply: a real person's web keeps taking in beliefs from perception, so it has to fit experience too.
Then reliabilism. Someone whose process is reliable, but who has no idea it's reliable, doesn't seem justified. The reply: a young child knows a parent's face without being able to defend the process, so demanding reasons you can recite sets the bar where the skeptic wants it.
That's three theories, three objections, and three replies, one of each per theory. Whether any of them also solves Gettier's problem is still open, and the course keys no verdict.
Pause and try this. What does the coherentist say to the worry about a made-up story?
Think about what perception adds.
Here's the answer. A real person's system keeps taking in beliefs from perception, which the world supplies, so a coherent system has to fit experience and can't drift off like a novel.
Closing
Let's pull it together. Justified true belief says knowing is believing something true for a good reason. Good reason isn't certainty, so argument 1 is valid but never keyed as sound. A definition claims every case, so 1 counterexample can defeat it. Gettier's 1963 paper offered cases of justified true belief that are true only by luck. In each one, what justifies the belief isn't what makes it true. A priori knowledge doesn't rest on experience, and that's about how, not when. The regress problem asks where reasons end, and foundationalism, coherentism, and reliabilism give rival answers.
Next in this module come the slide deck, the Readings and Resources page, the lecture tutorial, the practice set, the quiz, the discussion, and the assignment. In Week 6, The Mind-Body Problem, we ask what a mind is.
Built with The Course Maker · samples.thecoursemaker.com